Cover Image

You Can't Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectors

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Cao, Bochuan, Li, Changjiang, Cao, Yuanpu, Ge, Yameng, Wang, Ting, Chen, Jinghui
Format:	Preprint
Published:	2025
Subjects:	Cryptography and Security Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections
by: Cao, Yuanpu, et al.
Published: (2023)

Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM
by: Cao, Bochuan, et al.
Published: (2023)

Your Agent Can Defend Itself against Backdoor Attacks
by: Changjiang, Li, et al.
Published: (2025)

PRSA: Prompt Stealing Attacks against Real-World Prompt Services
by: Yang, Yong, et al.
Published: (2024)

Watch the Watcher! Backdoor Attacks on Security-Enhancing Diffusion Models
by: Li, Changjiang, et al.
Published: (2024)

Topology Matters: Measuring Memory Leakage in Multi-Agent LLMs
by: Liu, Jinbo, et al.
Published: (2025)

You Can Run But You Can't Hide: Runtime Protection Against Malicious Package Updates For Node.js
by: Ohm, Marc, et al.
Published: (2023)

Stealing User Prompts from Mixture of Experts
by: Yona, Itay, et al.
Published: (2024)

TruthFlow: Truthful LLM Generation via Representation Flow Correction
by: Wang, Hanyu, et al.
Published: (2025)

Prompt Pirates Need a Map: Stealing Seeds helps Stealing Prompts
by: Mächtle, Felix, et al.
Published: (2025)

Towards Robust Multimodal Large Language Models Against Jailbreak Attacks
by: Yin, Ziyi, et al.
Published: (2025)

Clone What You Can't Steal: Black-Box LLM Replication via Logit Leakage and Distillation
by: Gharami, Kanchon, et al.
Published: (2025)

Prompt Stealing Attacks Against Large Language Models
by: Sha, Zeyang, et al.
Published: (2024)

Prompt Leakage effect and defense strategies for multi-turn LLM interactions
by: Agarwal, Divyansh, et al.
Published: (2024)

RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction
by: Jiang, Tanqiu, et al.
Published: (2024)

Teach LLMs to Phish: Stealing Private Information from Language Models
by: Panda, Ashwinee, et al.
Published: (2024)

PARASITE: Conditional System Prompt Poisoning to Hijack LLMs
by: Pham, Viet, et al.
Published: (2025)

Intrinsic Fingerprint of LLMs: Continue Training is NOT All You Need to Steal A Model!
by: Yoon, Do-hyeon, et al.
Published: (2025)

GraphSteal: Structural Knowledge Stealing from Graph RAG via Traversal Reconstruction
by: Gu, Jinze, et al.
Published: (2026)

LeakDojo: Decoding the Leakage Threats of RAG Systems
by: Zhang, Maosen, et al.
Published: (2026)

You Can't Eat Your Cake and Have It Too: The Performance Degradation of LLMs with Jailbreak Defense
by: Mai, Wuyuao, et al.
Published: (2025)

Can You Keep a Secret? Involuntary Information Leakage in Language Model Writing
by: Holtzman, Ari, et al.
Published: (2026)

IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems
by: Wang, Liwen, et al.
Published: (2025)

PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt Sanitization
by: Geng, Runpeng, et al.
Published: (2025)

The Ethics of Interaction: Mitigating Security Threats in LLMs
by: Kumar, Ashutosh, et al.
Published: (2024)

InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models
by: Li, Hao, et al.
Published: (2024)

Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning
by: Ostermann, Simon, et al.
Published: (2024)

Quantifying Association Capabilities of Large Language Models and Its Implications on Privacy Leakage
by: Shao, Hanyin, et al.
Published: (2023)

Ignore This Title and HackAPrompt: Exposing Systemic Vulnerabilities of LLMs through a Global Scale Prompt Hacking Competition
by: Schulhoff, Sander, et al.
Published: (2023)

Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory
by: Mireshghallah, Niloofar, et al.
Published: (2023)

Harry Potter is Still Here! Probing Knowledge Leakage in Targeted Unlearned Large Language Models via Automated Adversarial Prompting
by: To, Bang Trinh Tran, et al.
Published: (2025)

What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks
by: Kirch, Nathalie, et al.
Published: (2024)

Fingerprinting LLMs via Prompt Injection
by: Hu, Yuepeng, et al.
Published: (2025)

Enhancing Prompt Injection Attacks to LLMs via Poisoning Alignment
by: Shao, Zedian, et al.
Published: (2024)

ContextLeak: Auditing Leakage in Private In-Context Learning Methods
by: Choi, Jacob, et al.
Published: (2025)

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs
by: Berezin, Sergey, et al.
Published: (2025)

When Agents "Misremember" Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems
by: Xu, Naen, et al.
Published: (2026)

Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data
by: Liang, Zi, et al.
Published: (2025)

Data Free Backdoor Attacks
by: Cao, Bochuan, et al.
Published: (2024)

Are All Prompt Components Value-Neutral? Understanding the Heterogeneous Adversarial Robustness of Dissected Prompt in Large Language Models
by: Zheng, Yujia, et al.
Published: (2025)